Correction for Partial Volume Effect Is a Must, Not a Luxury, to Fully Exploit the Potential of Quantitative PET Imaging in Clinical Oncology

(1)

Article

Reference

Correction for Partial Volume Effect Is a Must, Not a Luxury, to Fully Exploit the Potential of Quantitative PET Imaging in Clinical Oncology

ALAVI, Abass, et al.

Abstract

The partial volume effect (PVE) is considered as one of the major degrading factors impacting image quality and hampering the accuracy of quantitative PET imaging in clinical oncology.

This effect is the consequence of the limited spatial resolution of whole-body PET scanners, which results in blurring of the generated images by the scanner's response function. A number of strategies have been devised to deal with partial volume effect. However, the lack of consensus on the clinical relevance of partial volume correction and the most appropriate technique to be used in the context of clinical oncology limited their application in clinical setting. This issue is debated in this commentary.

ALAVI, Abass, et al . Correction for Partial Volume Effect Is a Must, Not a Luxury, to Fully

Exploit the Potential of Quantitative PET Imaging in Clinical Oncology. Molecular Imaging and Biology , 2018, vol. 20, no. 1, p. 1-3

PMID : 29181818

DOI : 10.1007/s11307-017-1146-y

Available at:

http://archive-ouverte.unige.ch/unige:99865

Disclaimer: layout of this document may differ from the published version.

1 / 1

(2)

Mol Imaging Biol (2017) DOI: 10.1007/s11307-017-1146-y

*World Molecular Imaging Society, 2017

COMMENTARY

Correction for Partial Volume Effect Is a Must, Not a Luxury, to Fully Exploit the Potential

of Quantitative PET Imaging in Clinical Oncology

Abass Alavi,

¹

Thomas J. Werner,

¹

Poul Flemming Høilund-Carlsen,

²

Habib Zaidi

^2,3,4,5

1Department of Radiology, Hospital of the University of Pennsylvania, 3400 Spruce Street, Philadelphia, PA, 19104, USA

2Department of Nuclear Medicine, University of Southern Denmark, DK-500, Odense, Denmark

3Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, 1211, Geneva, Switzerland

4Geneva Neuroscience Centre, University of Geneva, 1205, Geneva, Switzerland

5Department of Nuclear Medicine and Molecular Imaging, University of Groningen, Groningen, Netherlands

Abstract

The partial volume effect (PVE) is considered as one of the major degrading factors impacting image quality and hampering the accuracy of quantitative PET imaging in clinical oncology. This effect is the consequence of the limited spatial resolution of whole-body PET scanners, which results in blurring of the generated images by the scanner’s response function. A number of strategies have been devised to deal with partial volume effect. However, the lack of consensus on the clinical relevance of partial volume correction and the most appropriate technique to be used in the context of clinical oncology limited their application in clinical setting. This issue is debated in this commentary.

Key words: PET, Partial volume effect, Partial volume correction, Quantification, Clinical oncology

PET is without any doubt the imaging modality of choice for thein vivoquantitative assessment of molecular targets in a wide range of oncologic malignancies. However, its quantitative potential is still hampered by a number of physical degrading factors including partial volume effect, which results from the limited spatial resolution of current generation whole-body PET scanners. Although the total number of counts is preserved, partial volume effect spreads them over a larger region, thus making tiny malignant lesions appear greater but with less tracer uptake [1,2]. A number of partial-volume-correction (PVC) strategies have been devised and applied with a certain degree of success in research studies, albeit without reaching the level of maturity needed for implementation in the clinic, predominantly in oncological, cardiovascular, and neurological imaging where

it is expected to open new avenues in disease understanding and management [3].

The systematic review and meta-analysis by Cysouw et al. about the impact of PVC in quantification of malignant lesions is an important contribution to the field [4]. The authors have made an effort in reviewing the existing literature reaching a conclusion that the role of PVC on a routine basis in daily clinical practice of PET is at present still uncertain given the many uncertainties involved in PVC methodology. We wish to point out that regardless of the outcome of this meta-analysis, which intrinsically bears a number of limitations and pitfalls [5,6], the authors of the article above recognized in a previous article that ignoring such an important element in PET quantification may adversely affect the impact of this most powerful modality in managing patients with cancer and other serious disorders [7]. This study assessed the influence of PVC combined with several volumes of interest (VOI) techniques on the accuracy

Correspondence to:Abass Alavi;e-mail:abass.alavi@uphs.upenn.edu

(3)

and precision of PET quantification. In line with the meta- analysis, the authors suggest that PVC is beneficial but that several challenges prevent its use in clinical setting. A comprehensive discussion of the limitations of meta-analysis is beyond the scope of this communication. Still, we would like to express our opinion thatBthe assumption that a meta- analysis routinely represents the final and accurate view- point in an area of research is not warranted^[8].

Initial applications of tomographic imaging either with SPECT or PET primarily dealt with assessment of central nervous system and cardiac disorders. However, compared to SPECT, PET imaging has had a far-reaching impact in investigating a multitude of maladies throughout the body.

Significant improvement in spatial resolution of PET instruments over the past two decades has allowed visual- izing small lesions in many organs. Unfortunately, this has given the impression that images generated by PET can be used without further refinement, assuming they represent the ultimate results for both research and clinical purposes.

One of the early observations made when combining PET images with those of CT and MRI to determine the complementary roles of structural and molecular imaging was the impact of unavoidable and unexpected effects of structural abnormalities on quantitative measurements made by PET. Our group noted cortical atrophy as seen on CT or MRI results in significant underestimation of metabolic activity as calculated by PET [9]. This observation led to employing somewhat primitive approaches to segment the brain and the cerebral spinal fluid (CSF) to correct partial volume effects of brain activity and improve the quantification of the imaging data with PET. Soon thereafter, the introduction of novel segmentation techniques with MRI further enhanced the impact of PVC in assessing abnormalities in various organs [10]. This phenomenon is particularly true for abnormalities that are located in mid-trunk regions where the body size, juxtaposition of different tissue classes, and motion are major offenders for accurate quantification with PET.

Segmentation, although a major issue facing PET quantification, is a challenge in itself that is outside the scope of this comment. The basic physics of PET supports the idea that PVC of PET data contributes to the improvement of the quantitative accuracy of image-derived PET metrics and interpretive certainty. We agree that there is a way to go before the clinical significance of PVC is clearly documented. However, this does not justify explaining a way or minimizing the crucial influence that PVC has upon quantitative PET measurements. PVC is strongly needed to obtain reliable numbers for accumulated tracer activity in the body, not only in single PET scans, but particularly when repeat scans are required, because then, all measurement errors contribute twice, which puts even greater demand on measurement certainty and its critical role in assessing response to treatment when significant shrinkage and reduction in

tumor size is expected [11]. This might probably also apply to differentiation and better characterization of suspicious small focal uptake.

Ignoring PVC has resulted in advocating the view that lung lesions with uncorrected SUV less than 2.5 are benign and should not be investigated by surgery or biopsy for determining the true nature of such abnormalities. Unfortu- nately, this poorly conceived concept led to under- diagnosing this lethal cancer in many patients and its serious consequences [12–15]. These investigators were unaware that spatial resolution of PET in clinical imaging (lung) is substantially lower than what is measured when character- izing the performance of PET scanners using phantom studies [16, 17]. While the spatial resolution of PET as measured using point sources is approximately 4–6 mm, its true value in the trunk is substantially worse in organs such as the lungs and the liver, and as such, lesions that are smaller than 1 cm may pose a serious challenge with this technology. This is particularly true in lesions that are located in the lower lungs and in the liver where respiratory motion substantially blurs PET images [18, 19]. Unfortunately, respiratory motion correction is suboptimal for correcting this physiological artifact and has not been able to adequately address this major issue in the clinical setting.

The major advantages of PET over conventional imaging techniques including CT and MRI are its extraordinary sensitivity and potential accurate quantification of disease activity at various stages of the disease. Therefore, improve- ments in accurate quantification and efforts to perform global assessment of disease activity throughout the body in patients with cancer or other disorders have become a necessity for both diagnostic and therapeutic purposes [20].

Such measurements will require partial volume-corrected values along with accurate measurement of the size of the abnormalities visualized by PET [21]. This will allow for quantification of global disease activity in each lesion as well as the overall disease burden throughout the body.

These measurements will heavily depend on PVC, and therefore, ignoring this important element in PET quantification will undermine the impact of this incredibly powerful technology [22–24].

In recent years, great interest in assessing cardiac calcification in the coronary arteries has resulted in adopting measurement techniques that are suboptimal for detection and quantification of calcification in arteries that are only 1–

2 mm in diameter and are subject to cardiac and respiratory motion [25,26]. We believe every effort should be made to avoid such approaches in the future. Therefore, there is a dire need to employ methodologies that can overcome these unavoidable physical and physiological obstacles for accurate performance of PET imaging in certain organ disorders.

We agree with the authors about the difficulties that we are facing in accurate partial volume correction with the existing modalities used by practicing PET specialists. But we believe that what is currently being done is not adequate Alavi A. et al.: Correction for Partial Volume Effect Is a Must, Not a Luxury

(4)

enough for optimal utilization of the information that is provided by PET for managing the most serious disease of mankind. Therefore, we would recommend major efforts be made towards standardization and harmonization of imaging protocols, consensus on which PVC and consequent segmentation technique to be used for each specific application and localization, assessment of their characteristics, and clinical value on the way to their implementation on software supplied by scanner manufacturers for routine management of patients with cancer and other serious diseases.

Compliance with Ethical Standards Conflict of Interest

All authors declare that they have no conflict of interest.

References

1. Rousset O, Rahmim A, Alavi A, Zaidi H (2007) Partial volume correction strategies in PET. PET Clin 2(2):235–249.https://doi.org/

10.1016/j.cpet.2007.10.005

2. Soret M, Bacharach SL, Buvat I (2007) Partial-volume effect in PET tumor imaging. J Nucl Med 48(6):932–945.https://doi.org/10.2967/

jnumed.106.035774

3. Erlandsson K, Buvat I, Pretorius PH, Thomas BA, Hutton BF (2012) A review of partial volume correction techniques for emission tomography and their applications in neurology, cardiology and oncology. Phys Med Biol 57(21):R119–RR59. https://doi.org/

10.1088/0031-9155/57/21/R119

4. Cysouw MCF, Kramer GM, Schoonmade LJ, Boellaard R, de Vet HCW, Hoekstra OS (2017) Impact of partial-volume correction in oncological PET studies: a systematic review and meta-analysis. Eur J Nucl Med Mol Imaging 44(12):2105–2116.https://doi.org/10.1007/

s00259-017-3775-4

5. Flather MD, Farkouh ME, Pogue JM, Yusuf S (1997) Strengths and limitations of meta-analysis: larger studies may be more reliable.

Control Clin Trials 18(6):568–579; discussion 661-66.https://doi.org/

10.1016/S0197-2456(97)00024-X

6. Greco T, Zangrillo A, Biondi-Zoccai G, Landoni G (2013) Meta- analysis: pitfalls and hints. Heart Lung Vessel 5:219–225

7. Cysouw MC, Kramer GM, Hoekstra OS, Frings V, de Langen AJ, Smit EF et al (2016) Accuracy and precision of partial volume correction in oncological PET/CT studies. J Nucl Med 57(10):1642– 1649.https://doi.org/10.2967/jnumed.116.173831

8. Roehm E (2005) Limit ations of meta-analyses. http://

www.improvingmedicalstatistics.com/Limitations%20of%20Meta- analysis1.htm. 09/06/2017

9. Chawluk JB, Alavi A, Dann R, Hurtig HI, Bais S, Kushner MJ, Zimmerman RA, Reivich M (1987) Positron emission tomography in aging and dementia: effect of cerebral atrophy. J Nucl Med 28(4):431– 437

10. Tanna NK, Kohn MI, Horwich DN, Jolles PR, Zimmerman RA, Alves WM, Alavi A (1991) Analysis of brain and cerebrospinal fluid volumes with MR imaging: impact on PET data correction for atrophy. Part II. aging and Alzheimer dementia. Radiology 178(1):123–130.https://doi.org/10.1148/radiology.178.1.1984290 11. Hoilund-Carlsen PF, Lauritzen SL, Marving J, Rasmussen S, Hesse B,

Folke K, Godtfredsen J, Chraemmer-Jorgensen B, Gadsboll N, Dige- Petersen H (1988) The reliability of measuring left ventricular ejection fraction by radionuclide cardiography: evaluation by the method of variance components. Br Heart J 59(6):653–662. https://doi.org/

10.1136/hrt.59.6.653

12. Bar-Shalom R, Valdivia AY, Blaufox MD (2000) PET imaging in oncology. Semin Nucl Med 30(3):150–185.https://doi.org/10.1053/

snuc.2000.7439

13. Duhaylongsod FG, Lowe VJ, Patz EF Jr, Vaughn AL, Coleman RE, Wolfe WG (1995) Detection of primary and recurrent lung cancer by means of F-18 fluorodeoxyglucose positron emission tomography (FDG PET). J Thorac Cardiovasc Surg 110(1):130–139; discussion 39–40.https://doi.org/10.1016/S0022-5223(05)80018-2

14. Patz EF Jr, Lowe VJ, Hoffman JM, Paine SS, Burrowes P, Coleman RE, Goodman PC (1993) Focal pulmonary abnormalities: evaluation with F-18 fluorodeoxyglucose PET scanning. Radiology 188(2):487– 490.https://doi.org/10.1148/radiology.188.2.8327702

15. Prauer HW, Weber WA, Romer W, Treumann T, Ziegler SI, Schwaiger M (1998) Controlled prospective study of positron emission tomography using the glucose analogue [18f]

fluorodeoxyglucose in the evaluation of pulmonary nodules. Br J S ur g 8 5( 11 ): 15 06–15 11 . h t t p s : / / d o i . o r g / 1 0. 10 46 / j . 1 3 6 5 - 2168.1998.00915.x

16. Hickeson M, Yun M, Matthies A, Zhuang H, Adam LE, Lacorte L, Alavi A (2002) Use of a corrected standardized uptake value based on the lesion size on CT permits accurate characterization of lung nodules on FDG-PET. Eur J Nucl Med Mol Imaging 29(12):1639–1647.

https://doi.org/10.1007/s00259-002-0924-0

17. Kwee TC, Cheng G, Lam MG, Basu S, Alavi A (2013) SUVmax of 2.5 should not be embraced as a magic threshold for separating benign from malignant lesions. Eur J Nucl Med Mol Imaging 40:1475–1477 18. Salavati A, Borofsky S, Boon-Keng TK, Houshmand S, Khiewvan B, Saboury B, Codreanu I, Torigian DA, Zaidi H, Alavi A (2015) Application of partial volume effect correction and 4D PET in the quantification of FDG avid lung lesions. Mol Imaging Biol 17(1):140–148.https://doi.org/10.1007/s11307-014-0776-6

19. Teo BK, Saboury B, Munbodh R, Scheuermann J, Torigian DA, Zaidi H, Alavi A (2012) The effect of breathing irregularities on quantitative accuracy of respiratory gated PETCT. Med Phys 39(12):7390–7397.

https://doi.org/10.1118/1.4766876

20. Alavi A, Newberg AB, Souder E, Berlin JA (1993) Quantitative analysis of PET and MRI data in normal aging and Alzheimer’s disease: atrophy weighted total brain metabolism and absolute whole brain metabolism as reliable discriminators. J Nucl Med 34(10):1681– 1687

21. Berkowitz A, Basu S, Srinivas S, Sankaran S, Schuster S, Alavi A (2008) Determination of whole-body metabolic burden as a quantitative measure of disease activity in lymphoma: a novel approach with fluorodeoxyglucose-PET. Nucl Med Commun 29(6):521–526.https://

doi.org/10.1097/MNM.0b013e3282f813a4

22. Basu S, Zaidi H, Salavati A, Hess S, Carlsen PF, Alavi A (2014) FDG PET/CT methodology for evaluation of treatment response in lymphoma: from Bgraded visual analysis^ and Bsemiquantitative SUVmax^ to global disease burden assessment. Eur J Nucl Med Mol Imaging 41(11):2158–2160.https://doi.org/10.1007/s00259-014- 2826-3

23. Houshmand S, Salavati A, Hess S, Werner TJ, Alavi A, Zaidi H (2015) An update on novel quantitative techniques in the context of evolving whole-body PET imaging. PET Clin 10(1):45–58.https://

doi.org/10.1016/j.cpet.2014.09.004

24. Torigian DA, Lopez RF, Alapati S, Bodapati G, Hofheinz F, van den Hoff J, Saboury B, Alavi A (2011) Feasibility and performance of novel software to quantify metabolically active volumes and 3D partial volume corrected SUV and metabolic volumetric products of spinal bone marrow metastases on 18F-FDG-PET/CT. Hell J Nucl Med 14(1):8–14

25. Joshi NV, Vesey AT, Williams MC, Shah AS, Calvert PA, Craighead FH et al (2014) 18F-fluoride positron emission tomography for identification of ruptured and high-risk coronary atherosclerotic plaques: a prospective clinical trial. Lancet 383(9918):705–713.

https://doi.org/10.1016/S0140-6736(13)61754-7

26. Dweck MR, Aikawa E, Newby DE, Tarkin JM, Rudd JH, Narula J et al (2016) Noninvasive molecular imaging of disease activity in atherosclerosis. Circ Res 119(2):330–340. https://doi.org/10.1161/

CIRCRESAHA.116.307971

Alavi A. et al.: Correction for Partial Volume Effect Is a Must, Not a Luxury